DESCRIPTION: Despite taking daily medications, over 1/3 of patients with epilepsy continue to have seizures, and many more experience significant adverse side effects from antiepileptic drugs (AED). Though patients may spend as little as 0.01% of their time actually having seizures, they must maintain chronic therapeutic AED levels for years to decades. The reason AEDs are taken daily for seizures that only occur sporadically, and in many cases infrequently, is because patients do not know when a seizure will strike. Arguably the unpredictability of seizures and the adverse cognitive and physical side effects associated with AEDs, are the most disabling aspects of epilepsy. We now have compelling evidence from humans that seizures are not random events, but rather are associated with intracranial EEG (lEEG) changes that occur tens of minutes to hours before clinical events. This research has lead NeuroVista Inc. to develop a seizure advisory system (SAS) that uses iEEG and machine learning algorithms to track the probability of seizure occurrence. Using patient specific algorithms NeuroVista's SAS can forecast periods of increased and decreased seizure likelihood with high sensitivity and specificity, benchmarked against a chance predictor. NeuroVista's SAS has been validated in humans undergoing IEEG monitoring for epilepsy surgery, and has recently been approved for a human pilot clinical trial in Australia. We propose to use the NeuroVista's SAS device to guide the delivery of AEDs. We hypothesize that intelligent delivery of AEDs at times of increased seizure likelihood will effectively prevent clinical seizures. During periods of low seizure likelihood AEDs would not be required, reducing side effects. In this proposal we develop and test neurophysiologically-based responsive pharmacotherapy in naturally occurring canine epilepsy as the first step in creating a new treatment paradigm for medically resistant human partial epilepsy. Naturally occurring canine partial epilepsy is an ideal model for developing this therapy because of the clinical, physiological, and pharmacological similarities between canine and human partial epilepsy. In addition, dogs are large enough to tolerate implantation of the human SAS device. This project will lead to submission of an investigational device exemption to conduct a first-in-human clinical trial of SAS guided responsive AED therapy. Disclaimer: Please note that the following critiques were prepared by the reviewers prior to the Study Section meeting and are provided in an essentially unedited form. While there is opportunity for the reviewers to update or revise their written evaluation, based upon the group's discussion, there is no guarantee that individual critiques have been updated subsequent to the discussion at the meeting. Therefore, the critiques may not fully reflect the final opinions of th individual reviewers at the close of group discussion or the final majority opinion of the group. Thus the Resume and Summary of Discussion is the final word on what the reviewers actually considered critical at the meeting.